CN107632225A - A kind of small current system Earth design method - Google Patents
A kind of small current system Earth design method Download PDFInfo
- Publication number
- CN107632225A CN107632225A CN201710687706.3A CN201710687706A CN107632225A CN 107632225 A CN107632225 A CN 107632225A CN 201710687706 A CN201710687706 A CN 201710687706A CN 107632225 A CN107632225 A CN 107632225A
- Authority
- CN
- China
- Prior art keywords
- fault
- feeder
- curve
- characteristic value
- data window
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Locating Faults (AREA)
Abstract
The invention discloses a kind of small current system Earth design method, including：Extract each zero-sequence current of the line feed terminals before and after fault moment on fault feeder；Point-by-point shifting data window simultaneously calculates the characteristic value of zero-sequence current of each line feed terminals under each data window；Build the eigenvalue graph of each line feed terminals；Obtain sampled point corresponding with eigenvalue of maximum in each eigenvalue graph for the line feed terminals that bus is nearest on fault feeder；Obtain characteristic value corresponding with the identical sampled point of above-mentioned acquisition in the eigenvalue graph of remaining line feed terminals；Calculate the Euclidean distance of the characteristic value of two neighboring line feed terminals；By Euclidean distance according to arrange from big to small and choose maximum N number of Euclidean distance；If an Euclidean distance is more than remaining 1 Euclidean distance sum of N, the section between two line feed terminals corresponding to the Euclidean distance is fault section.The present invention can improve the reliability of fault location and mitigate the communication pressure of data channel by the above method.
Description
Technical Field
The invention relates to the technical field of power distribution network fault detection, in particular to a method for positioning a ground fault of a low-current system.
Background
In China, the area distribution of a power distribution network is wide, the fault rate is high, and the protection configuration is relatively low. The medium-voltage distribution network is mainly based on an overhead line, the proportion of single-phase earth faults is large, and because the distribution network generally adopts a low-current earthing mode, a short circuit loop is not formed during the single-phase earth faults, and only a low fault current is generated by a zero-sequence capacitor of a feeder line, so that the system can still continue to operate with the faults for 1-2 hours. In order to avoid further expansion of the single-phase earth fault, an earth fault point needs to be found as soon as possible to remove the fault. China has made a great deal of research on the small-current single-phase earth fault line selection technology, but the small-current earth fault location technology is relatively few in design, and the practical application effect is not ideal, so that the traditional manual line patrol method or line pull method is still widely adopted to determine the fault point on site at present, which not only consumes a great amount of manpower and material resources and causes great economic loss to users, but also reduces the power supply stability of the power distribution system, and is not beneficial to the development of power distribution automation.
The single-phase earth fault section positioning method generally comprises an impedance method, a zero sequence current amplitude comparison method, a 5-order harmonic method, a traveling wave method, a signal injection method, a correlation analysis method and the like. The impedance method is generally used for fault location of a high-voltage transmission line with a grounded neutral point, and has low availability in a power distribution network. The zero sequence current amplitude comparison method is suitable for a power distribution network system with an ungrounded neutral point, but the method is ineffective in positioning for the power distribution network with the grounded arc suppression coil. The 5 th harmonic method is affected by the network structure of the system, the grounding resistance and the fault closing angle, so that the fault positioning stability is poor. The traveling wave method generally requires a very high sampling frequency, and meanwhile, the power distribution network has a complex structure with multiple feeders and multiple branches, so that the traveling wave can generate multiple refraction and reflection waves. Although the signal injection method has certain advantages, the strength of the injected signal is limited by the capacity of a Potential Transformer (PT), and additional signal generating equipment is required, which increases the investment cost and is relatively poor in economy. The related analysis method is used for positioning a fault section by comparing waveform similarity of zero sequence current, a large amount of sampling data is required to be uploaded, the requirement on signal synchronism is high, and communication pressure is increased. Artificial intelligence algorithms, such as neural networks, expert systems, etc., are also beginning to be applied to the research of single-phase earth fault location, but the calculation amount is large and the feasibility is low. The principle of fault section positioning is to realize fault positioning according to the difference of fault signals of a fault upstream feeder terminal and a fault downstream feeder terminal.
Aiming at the defects that the existing fault positioning method (such as a correlation analysis method) needs a feeder terminal to upload one or more cycles of sampling data, the sampling frequency is generally kilohertz, which causes certain pressure on communication, and the existing fault positioning method (such as a 5 th harmonic method) only uses single fault information, has less fault feature extraction, low fault positioning reliability and smaller practical application range, and the small-current fault positioning method which can realize high fault positioning reliability and small communication pressure is necessary to be provided.
Disclosure of Invention
Aiming at the defects of low positioning reliability and large data uploading amount of the existing small current fault positioning method, the small current system ground fault positioning method is actually needed to be provided, so that the data uploading amount is reduced, the communication pressure of a data channel is reduced, and the positioning reliability is improved.
In one aspect, the present invention provides a method for positioning a low current fault, including:
step 1: extracting zero sequence currents of each feeder line terminal on a fault feeder line before and after the fault moment;
acquiring zero-sequence voltage of a bus in real time, determining a fault feeder line according to a fault line selection method when the zero-sequence voltage of the bus is detected to exceed a fault judgment threshold, and extracting zero-sequence current of P power frequency periods before a fault moment and Q power frequency periods after the fault moment of each feeder line terminal in M feeder line terminals on the fault feeder line;
wherein P is more than or equal to 0.5 and less than or equal to 2, Q is more than or equal to 1 and less than or equal to 3, and M is more than or equal to 3;
step 2: selecting the length of a data window as a power frequency cycle, moving the data window point by point on the zero sequence current of each feeder line terminal extracted in the step 1, and calculating the characteristic value of the zero sequence current of each feeder line terminal under each data window;
the characteristic values comprise three types of characteristic values of skewness, variance and extreme values, and each feeder terminal is provided with D sampling points in P + Q power frequency periods;
and step 3: constructing a characteristic value curve of each feeder terminal according to the characteristic values calculated in the step 2;
the characteristic value curve comprises a skewness curve, a variance curve and an extreme value curve, and is a curve of a characteristic value-sampling point;
sampling points in the characteristic value curve are sampling points in the first P + Q-1 power frequency cycles in the zero sequence current of the feeder terminal extracted in the step 1;
and 4, step 4: acquiring a sampling point corresponding to the maximum characteristic value in each characteristic value curve of a feeder terminal closest to a bus on a fault feeder;
and 5: acquiring various characteristic values corresponding to the same sampling points in the step 4 in characteristic value curves of the remaining M-1 feeder terminals on the fault feeder;
and 6: sequentially calculating Euclidean distances of the characteristic values of two adjacent feeder line terminals according to the maximum characteristic value in the step 4 and the characteristic value obtained in the step 5;
and 7: arranging the Euclidean distances calculated in the step 6 from large to small, and selecting the maximum N Euclidean distances from the Euclidean distances;
wherein N is more than or equal to 3 and less than or equal to M-1;
and 8: if the Euclidean distance of the characteristic value of one adjacent two feeder line terminals is larger than the sum of the remaining N-1 Euclidean distances in the step 7, the interval between the two adjacent feeder line terminals corresponding to the Euclidean distance of the characteristic value of the one adjacent two feeder line terminals is a fault interval; and if not, returning to the step 1 to reselect the fault feeder line for fault location.
The corresponding relation between any one characteristic value in the characteristic value curve and the sampling point is as follows: and the characteristic values corresponding to all data of the data window with the sampling point as the starting point are used for the data window which moves point by point.
Preferably, the calculation formula of the deviation in the characteristic value in step 2 is as follows:
wherein, sk (i) 0ab (j) ) represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab Skewness of (2);
representing a random variable ofThe desired operator of (a);
i 0ab (n) is the zero sequence current sampled by the feeder terminal numbered b on the fault feeder a at the sampling point n, n is the sampling point, ns is the number of the sampling points under the selected data window length, mu and sigma respectively represent i 0ab Expectation and standard deviation of.
Preferably, the variance calculation formula in the feature values in step 2 is as follows:
Va(i 0ab (j))＝E[(i 0ab (n)-E(i 0ab (n))) 2 ],n＝j,j+1,...,j+Ns-1
wherein, va (i) 0ab (j) ) represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab The variance of (a);
E[(i 0ab (n)-E(i 0ab (n))) 2 ]represents a random variable of (i) 0ab (n)-E(i 0ab (n))) 2 The desired operator of (a); e (i) 0ab (n)) represents a random variable of i 0ab (n) a desired operator;
i 0ab and (n) is the zero sequence current collected by the feeder terminal with the number b on the fault feeder a at a sampling point n, wherein n is the sampling point, and Ns is the number of the sampling points under the selected data window length.
Preferably, the extreme value calculation formula in the characteristic value in step 2 is as follows:
Ra(i 0ab (j))＝max(i 0ab (n))-min(i 0ab (n)),n＝j,j+1,...,j+Ns-1
wherein, ra (i) 0ab (j) ) represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab The extreme value of (c);
max(i 0ab (n)) represents the zero sequence current i in the data window containing Ns sampling point numbers 0ab The maximum value of (a);
min(i 0ab (n)) represents the zero sequence current i in the data window containing Ns sampling point numbers 0ab N is the number of samples, and Ns is the number of samples at the selected data window length.
Preferably, the euclidean distance calculation formula of the feature values of two adjacent feeder terminals in step 6 is as follows:
wherein d (k, k-1) represents the Euclidean distance of characteristic values of the kth feeder terminal and the kth-1 feeder terminal;
Sk k 、Va k 、Ra k respectively representing a skewness curve and a variance curve of the kth feeder terminal and a skewness, a variance and an extreme value corresponding to the same sampling point in the step 4 on an extreme value curve; sk k-1 、Va k-1 、Ra k-1 And respectively representing the skewness curve, the variance curve and the extreme value corresponding to the same sampling point in the step 4 on the skewness curve, the variance curve and the extreme value curve of the k-1 th feeder terminal.
Preferably, N of the maximum N euclidean distances selected in step 7 is 3.
Has the beneficial effects that:
the invention provides a method for positioning a ground fault of a small current system, which comprises the steps of collecting zero sequence currents before and after a fault of a feeder terminal on a fault feeder, extracting characteristic values of the zero sequence currents, sequentially calculating Euclidean distances of the characteristic values of adjacent feeder terminals, identifying whether one Euclidean distance is larger than the sum of the remaining N-1 Euclidean distances, and if yes, determining that an interval between two adjacent feeder terminals corresponding to the one Euclidean distance is a fault interval. The fault location method is higher in reliability compared with the method that the fault location is carried out by only utilizing single fault information in the existing fault location, and in addition, the method extracts a plurality of characteristic values, so that the extracted characteristic values are only needed to be uploaded, the sampling data of the whole power frequency period are not needed to be uploaded, the uploaded data are greatly reduced, and the communication pressure of a data transmission channel is reduced.
In addition, the fault positioning method of the invention realizes fault positioning by utilizing Euclidean distances of a plurality of characteristic values of two adjacent feeder line terminals, and has no requirement on time synchronization of uploaded data, namely the fault positioning method does not need strict communication synchronization.
The method selects the feeder terminal closest to the bus as a reference, acquires the sampling points corresponding to the maximum characteristic values in each characteristic value curve, and acquires the corresponding characteristic values in each characteristic value curve of the rest fault terminals according to the sampling points.
Moreover, the invention analyzes a plurality of characteristic values rather than all sampling data, the data calculation amount is small, and the invention is not only suitable for a neutral point ungrounded system, but also suitable for a neutral point arc suppression coil grounded system, and has strong practicability.
Drawings
Fig. 1 is a schematic flow chart of a method for locating a ground fault of a low-current system according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a method for locating a ground fault of a low current system according to an embodiment of the present invention;
fig. 3 is a zero sequence network equivalent circuit diagram of a single-phase earth fault provided by an embodiment of the present invention;
fig. 4 is a characteristic graph provided by an embodiment of the present invention, in which (a) is a skewness graph, (b) is a variance graph, and (c) is an extremum graph.
Detailed Description
The invention is further described below with reference to the drawings and the detailed description.
Referring to fig. 1, the method for locating a ground fault of a low current system provided by the invention comprises the following steps:
step 1: extracting zero sequence currents of each feeder line terminal on a fault feeder line before and after the fault moment;
the method comprises the steps of collecting zero sequence voltage of a bus in real time, starting a fault positioning algorithm when the zero sequence voltage of the bus is detected to exceed a fault discrimination threshold, determining a fault feeder line according to a fault line selection method, and extracting zero sequence current i of each feeder line terminal in M feeder line terminals on the fault feeder line in P power frequency periods before a fault moment and Q power frequency periods after the fault moment 0ab (n)；
Wherein i 0ab And (n) is expressed as a zero sequence current value of the b-th feeder terminal on the a-th feeder at a sampling point n, wherein P is more than or equal to 0.5 and less than or equal to 2, Q is more than or equal to 1 and less than or equal to 3, M is more than or equal to 3, and P =1.5 and Q =2.5 are preferred in the embodiment.
Specifically, the bus zero sequence voltage u is acquired in real time by using a bus voltage sensor 0 (n), where the fault determination threshold is a multiplication value of a reliability coefficient and a rated voltage, the zero-sequence voltage caused by the unbalanced voltage of the system is considered to be avoided in this embodiment, and the reliability coefficient is generally 0.3, as follows:
u 0 (n)＞K e U N
wherein u is 0 (n) is the bus zero sequence voltage at sampling point n, K e For a reliability factor, U N Is the rated voltage.
It should be understood that when the instantaneous value of the zero sequence voltage of the bus is greater than the fault judgment threshold value, the system is judged to have a single-phase earth fault and the fault time can be known. The fault line selection method adopted in the embodiment is implemented based on the existing line selection method, such as a amplitude comparison line selection method, a phase comparison line selection method, an injection method and a correlation analysis method.
The Feeder line is provided with M Feeder line terminals (FTU), the Feeder line terminals are used for collecting zero sequence current of a measuring point in real time, the ith Feeder line Terminal is represented by FTUi, and i is more than or equal to 1 and less than or equal to M. Fig. 3 is an equivalent circuit diagram of a simulation model of a small current ground fault of a 110kV/10kV distribution network, wherein 4 feeder lines are provided, each feeder line is provided with a corresponding FTU, and the feeder line lengths are respectively 5km,7km,10km and 12km, wherein the feeder line 3 has a single-phase ground fault and is located between the FTU2 and the FTU3, the grounding mode of a neutral point is controlled by a switch K, the switch is disconnected into a neutral point non-grounding mode, and is closed into an arc suppression coil grounding mode, the feeder lines in the diagram are all overhead lines, and zero-sequence parameters are R0=0.25 Ω/km, L0=5.54mH/km, and C0=0.011 μ F/km; the positive sequence parameters are R1=0.178 Ω/km, L1=1.21mH/km, C1=0.012 μ F/km, and the system sampling frequency is 2kHz.
Step 2: selecting the length of a data window as a power frequency cycle, moving the data window point by point on the zero sequence current of each feeder line terminal extracted in the step 1, and calculating the characteristic value of the zero sequence current of each feeder line terminal under each data window;
the characteristic values comprise three types of characteristic values of skewness, variance and extreme values, and each feeder terminal is provided with D sampling points in P + Q power frequency periods.
Specifically, if the feeder terminal has Ns sampling points in a power frequency cycle, then a data window contains the zero sequence current of Ns sampling points, then in step 2, according to the zero sequence current of the feeder terminal in P + Q power frequency cycles extracted in step 1, the characteristic value of the sampling point in a power frequency cycle with the first sampling point as the starting point is calculated, wherein Ns sampling points are in a data window, then one sampling point is moved backwards, the characteristic values of the Ns sampling points in a data window with the sampling point as the starting point are calculated, the data window is moved backwards point by point all the time, and the characteristic value corresponding to the zero sequence current of the moved data window is calculated each time the data window is moved. It should be understood that the number of times the data window is moved is equal to the number of sampling points in the previous P + Q-1 power frequency cycles in the data of P + Q power frequency cycles collected in step 1. For example, in step 1, P =1.5, q =2.5, then the number of times of moving the data window is equal to the number of sampling points in the first 3 power frequency cycles.
Wherein, the calculation formula of the skewness of the zero sequence current in the selected data window is as follows:
wherein, sk (i) 0ab (j) Represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab The skewness of (2).
i 0ab For the zero sequence current sampled by the feeder terminal numbered b on the faulty feeder a,representing a random variable ofThe desired operator of (a);
i 0ab (n) is the zero sequence current sampled by the number b of the feeder terminal on the fault feeder a at the sampling point n, n is the sampling point, j is also the sampling point, ns is the number of the sampling points under the selected data window length, mu and sigma respectively represent i 0ab Expectation and standard deviation of. .
The calculation formula for the above segment is further developed as follows:
in this embodiment, j is specifically represented as a sampling point in the first P + Q-1 power frequency cycles in the collected data of the P + Q power frequency cycles. It should be understood that j and n both represent sample points.
The formula for calculating the variance of the zero sequence current in the selected data window is as follows:
Va(i 0ab (j))＝E[(i 0ab (n)-E(i 0ab (n))) 2 ],n＝j,j+1,...,j+Ns-1
wherein, va (i) 0ab (j) Represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab The variance of (a); e2(i 0ab (n)-E(i 0ab (n))) 2 ]Represents a random variable of (i) 0ab (n)-E(i 0ab (n))) 2 The desired operator of (a); e (i) 0ab (n)) represents a random variable of i 0ab (n) a desired operator; i.e. i 0ab And (n) is zero sequence current collected by a feeder terminal with the number of b on the fault feeder a at a sampling point n, wherein n is the sampling point, and Ns is the number of the sampling points under the selected data window length.
It should be understood that i 0ab N in (n) and E (i) 0ab N in (n)) are not synchronously equal, e.g. when i is equal 0ab N = j =1 in (n), E (i) 0ab (n)) has a value of i 0ab (1)、i 0ab (2)…i 0ab (Ns) desired operator, then E (i) 0ab (n)) wherein n is 1, 2.
The formula for calculating the extreme value of the zero-sequence current in the selected data window is as follows:
Ra(i 0ab (j))＝max(i 0ab (n))-min(i 0ab (n)),n＝j,j+1,...,j+Ns-1
wherein, ra (i) 0ab (j) Represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab The extreme value of (c);
max(i 0ab (n)) represents the zero sequence current i in the data window containing Ns sampling points 0ab The maximum value of (a);
min(i 0ab (n)) represents the zero sequence current i in the data window containing Ns sampling point numbers 0ab N is the number of samples, and Ns is the number of samples at the selected data window length.
And 3, step 3: constructing a characteristic value curve of each feeder terminal according to the characteristic values calculated in the step 2;
the characteristic value curves include a skewness curve, a variance curve and an extremum curve, that is, each feeder terminal in this embodiment includes three characteristic value curves, which are the skewness curve, the variance curve and the extremum curve.
The characteristic value curve is a characteristic value-sampling point curve, and the sampling points in the characteristic curve are sampling points j in the first P + Q-1 power frequency cycles in the zero-sequence current of the feeder terminal extracted in the step 1.
The corresponding relation between any one characteristic value in the characteristic value curve and the sampling point is as follows: and the characteristic value corresponding to the data window with the sampling point j as the starting point is used for the data window which moves point by point.
In this embodiment, as shown in (a), (b), and (c) of fig. 4, the graph (a) is a skewness curve of one feeder terminal, the graph (b) is a variance curve of one feeder terminal, and the graph (c) is an extremum curve of one feeder terminal. Wherein the vertical axis is the eigenvalues and the horizontal axis is the sampling point. In this embodiment, step 1 acquires data of 4 power frequency periods, each power frequency period includes 40 sampling points, and as can be seen from the graphs (a), (b), and (c), a sampling point k on the horizontal axis is a sampling point corresponding to 3 power frequency periods, that is, a sampling point in the first 3 power frequency periods in the 4 power frequency periods acquired in step 1 in this embodiment, where the sampling point k on the horizontal axis = a characteristic value of a first data window corresponding to 1, that is, the data window has not moved yet; and the sampling point k =2 corresponds to the characteristic value of the second data window, the sampling point k =3 corresponds to the characteristic value of the third data window, and the analogy is repeated to obtain the characteristic values corresponding to the data windows with 120 sampling points as the starting points in the first 3 power frequency periods.
And 4, step 4: and acquiring a sampling point corresponding to the maximum characteristic value in each characteristic value curve of the feeder terminal closest to the bus on the fault feeder.
In this embodiment, the sampling points corresponding to the maximum characteristic values in the skewness curve, the variance curve and the extremum curve of the feeder terminal closest to the bus are respectively represented as j sk ，j Va ，j Ra 。
It should be noted that in some embodiments, the maximum feature value corresponds to a plurality of sample points, such as j sk For multiple values, in other embodiments, the maximum feature value corresponds to only one sample point.
For example, the feeder terminals corresponding to (a), (b), and (c) in fig. 4 are the feeder terminals closest to the bus, and as shown in the figure, the sampling point j corresponding to the maximum deviation in the graph (a) sk 22, sample point j corresponding to the maximum variance in graph (b) Va 61, maximum j corresponding to the extreme value in graph (c) Ra Sample points were 29-64.
And 5: obtaining various characteristic values corresponding to the same sampling points in the step 4 in the characteristic value curves of the rest M-1 feeder line terminals on the fault feeder line;
specifically, j in a characteristic value curve of the remaining M-1 feeder terminals is obtained sk ，j Va ，j Ra Corresponding skewness, variance, and extremum.
If the maximum eigenvalue obtained in step 4 corresponds to multiple sampling points, and when multiple eigenvalues correspond to the multiple sampling points in the eigenvalue curves of the remaining M-1 feeder terminals, in this embodiment, it is preferable to randomly select one eigenvalue from the multiple eigenvalues, in other possible embodiments, it may be possible to select an intermediate value from the multiple eigenvalues, or calculate an average value of the multiple eigenvalues, which is not specifically limited in the present invention.
Step 6: sequentially calculating Euclidean distances of the characteristic values of the two adjacent feeder line terminals according to the maximum characteristic value in the step 4 and the characteristic value obtained in the step 5.
The Euclidean distance calculation formula of the characteristic values of two adjacent feeder line terminals is as follows:
wherein d (k, k-1) represents a Euclidean distance of characteristic values of a kth feeder terminal and a k-1 th feeder terminal, wherein the kth feeder terminal and the k-1 th feeder terminal are adjacent feeder terminals.
Sk k 、Va k 、Ra k Respectively is the same sampling point j in the k-th feeder terminal on the skewness curve, the variance curve and the extreme value curve as the sampling point j in the step 4 sk ，j Va ，j Ra Corresponding skewness, variance and extreme value; sk k-1 、Va k-1 、Ra k-1 Respectively the skewness curve, variance curve and extreme value curve of the kth-1 th feeder terminal are the same as the sampling point j in the step 4 sk ，j Va ，j Ra Corresponding skewness,Variance, extremum.
It should be understood that in the present embodiment, if there are M feeder terminals, there are M-1 euclidean distances correspondingly.
And 7: arranging the Euclidean distances calculated in the step 6 from large to small, and selecting the maximum N Euclidean distances from the Euclidean distances;
wherein, N is more than or equal to 3 and less than or equal to M-1, N is preferably 3 in the embodiment, and the error is the smallest.
And 8: if the Euclidean distance of the characteristic value of one adjacent two feeder terminals is larger than the sum of the Euclidean distances of the characteristic values of the remaining N-1 adjacent two feeder terminals in the step 7, the interval between the two adjacent feeder terminals corresponding to the Euclidean distance of the characteristic value of one adjacent two feeder terminals is a fault interval; and if not, returning to the step 1 to reselect the fault feeder line for fault location.
Specifically, the following formula should be satisfied:
wherein d (i, i + 1) represents the Euclidean distance of the characteristic values of the ith feeder terminal and the (i-1) th feeder terminal which are adjacent.
As shown in fig. 2, in this embodiment, data of zero-sequence currents of 4 power frequency cycles are selected for analysis, and N =3 is selected for fault location, and if N =3, the 3 euclidean distances acquired in step 7 are d (r, r-1), d (s, s-1), d (t, t-1), and d (r, r-1) > d (s, s-1) > d (t, t-1), respectively;
if the relationship is satisfied:
d(r,r-1)＞d(s,s-1)+d(t,t-1)
the section between the r-th feeder terminal and the r-1 st feeder terminal is a faulty section.
The embodiment of the invention takes the model shown in fig. 3 as an example, and discusses two modes of neutral point ungrounded and neutral point grounded through an arc suppression coil respectively:
when the grounding mode of the neutral point is that the neutral point is not grounded, the feeder line 3 has a single-phase grounding fault and is positioned between the FTU2 and the FTU3, and four feeder line terminals are arranged on the feeder line 3 and are respectively represented as FTU1, FTU2, FTU3 and FTU4.
After the system has single-phase earth fault, extracting zero sequence currents (including i) of 4 power frequency periods of 4 feeder terminals of the feeder 3 (1.5T before fault and 2.5T after fault) 031 (n)、i 032 (n)、i 033 (n)、i 034 (n)) extracting characteristic values, wherein n is a sampling point;
and moving the data window with the length of one power frequency cycle point by point to obtain 3 characteristic value curves of the data extracted by the FTU1, the FTU2, the FTU3 and the FTU4, namely a skewness curve, a variance curve and an extreme value curve.
With FTU1 of the fault feeder line as reference, respectively obtaining the maximum value Max (Sk) of skewness, variance and extreme value curve 1 )、Max(Va 1 )、Max(Ra 1 ) Finding 3 sampling points corresponding to the maximum value, and setting the sampling points as j sk ，j Va ，j Ra . The skewness, variance and extremum (eigenvalue) curves of FTU1 are shown in (a), (b) and (c) of fig. 4, max (Sk) 1 )＝6.085；Max(Va 1 )＝92.83；Max(Ra 1 )＝40.84；j sk ＝22；，j Va ＝61；29<＝j Ra <＝64；
Solving j of FTU2, FTU3 and FTU4 according to characteristic value curves of FTU2, FTU3 and FTU4 sk ，j Va ，j Ra Corresponding skewness, variance and extremum. They are respectively:
Sk 2 (j sk )＝6.085；Va 2 (j Va )＝113.4；Ra 2 (j Ra )＝45.50；
Sk 3 (j sk )＝-6.085；Va 3 (j Va )＝4.417；Ra 3 (j Ra )＝9.664；
Sk 4 (j sk )＝-6.085；Va 4 (j Va )＝0.723；Ra 4 (j Ra )＝3.631；
respectively solving 3 of FTU1 and FTU2, FTU2 and FTU3, FTU3 and FTU4 according to the calculation formula of Euclidean distanceEuclidean distance d of characteristic value 12 、d 23 、d 34 Respectively is as follows:
d 12 ＝21.091；d 23 ＝115.37；d 34 ＝7.0741；
from the above, there is d 23 >d 12 +d 34 Therefore, the section between FTU2 and FTU3 is determined as the fault section.
Similarly, when the neutral grounding mode is grounding through the arc suppression coil, the feeder 3 has a single-phase grounding fault and is located between the FTU2 and the FTU3, and four feeder terminals, which are respectively represented as FTU1, FTU2, FTU3 and FTU4, are installed on the feeder 3.
After the system has single-phase earth fault, 4 FTUs of the feeder line 3 extract zero-sequence currents (including i) of 4 periods of 1.5T before fault and 2.5T after fault 031 (n)、i 032 (n)、i 033 (n)、i 034 (n)) extracting characteristic values;
respectively solving 3 characteristic value curves of data extracted by FTU1, FTU2, FTU3 and FTU4 by moving the data window point by point;
with FTU1 of the fault feeder line as reference, respectively obtaining the maximum value Max (Sk) of skewness, variance and extreme value curve 1 )、Max(Va 1 )、Max(Ra 1 ) Finding 3 sampling points corresponding to the maximum value of the characteristic value in the characteristic value curve, and setting the sampling points as j sk ，j Va ，j Ra 。
Max(Sk 1 )＝6.085；Max(Va 1 )＝91.05；Max(Ra 1 )＝39.65；j sk ＝22；j Va ＝61；29<＝j Ra <＝64；
According to characteristic value curves of FTU2, FTU3 and FTU4, j of FTU2, FTU3 and FTU4 is calculated sk ，j Va ，j Ra Corresponding skewness, variance and extremum. They are respectively:
Sk 2 (j sk )＝6.085；Va 2 (j Va )＝110.6；Ra 2 (j Ra )＝42.99；
Sk 3 (j sk )＝-6.085；Va 3 (j Va )＝4.417；Ra 3 (j Ra )＝9.660；
Sk 4 (j sk )＝-6.085；Va 4 (j Va )＝0.723；Ra 4 (j Ra )＝3.630；
respectively solving Euclidean distances d of 3 characteristic values of FTU1 and FTU2, FTU2 and FTU3 and FTU4 according to Euclidean distance formula 12 、d 23 、d 34 Respectively as follows:
d 12 ＝19.833；d 23 ＝111.95；d 34 ＝7.0715；
from the above, there are d 23 >d 12 +d 34 Therefore, the section between FTU2 and FTU3 is determined as the fault section.
In summary, compared with the existing fault positioning method, the method for positioning the ground fault of the small current system has the advantages that: the uploaded data is greatly reduced, and the communication pressure of a data transmission channel is reduced; the Euclidean distance of a plurality of characteristic values of two adjacent FTUs is utilized to realize fault positioning, and no requirement is placed on time synchronization of uploaded data, namely the fault positioning method does not need strict communication synchronization; the characteristic values can reflect fault information comprehensively, can realize fault positioning under different fault states, and has high reliability.
Specifically, the traditional method for positioning the ground fault of the small current system needs to upload the zero sequence current measured by each FTU in real time, so that the amount of uploaded data is large, and a traditional data channel is easy to block. Generally, a small-current grounding system has small fault steady-state current and rich high-frequency transient components, in order to better utilize the high-frequency transient components to perform fault location, the sampling frequency of the system needs to be increased, and the higher the sampling frequency is, the more sampling points are in the same time, and the larger the communication pressure is; if the number of the feeders and the FTUs is large, the upload data and the communication pressure are further increased, which causes communication malfunction, data loss, and the like, and finally causes failure of the fault location method. According to the method, only 3 characteristic values are required to be uploaded, so that the uploaded data are greatly reduced, and the communication pressure of a data transmission channel is reduced.
The foregoing is merely a preferred embodiment of the invention, which is intended to be illustrative and not limiting. The skilled person will understand that many modifications may be made thereto within the scope of the invention as defined in the claims, but all will fall within the scope of the invention.
Claims (6)
1. A method for locating a ground fault in a low current system, comprising:
step 1: extracting zero sequence current of each feeder terminal on a fault feeder before and after the fault moment;
acquiring zero-sequence voltage of a bus in real time, determining a fault feeder line according to a fault line selection method when the zero-sequence voltage of the bus is detected to exceed a fault judgment threshold, and extracting zero-sequence current of P power frequency cycles before a fault moment and Q power frequency cycles after the fault moment of each feeder line terminal in M feeder line terminals on the fault feeder line;
wherein P is more than or equal to 0.5 and less than or equal to 2, Q is more than or equal to 1 and less than or equal to 3, and M is more than or equal to 3;
step 2: selecting the length of a data window as a power frequency period, moving the data window point by point on the zero sequence current of each feeder terminal extracted in the step 1, and calculating the characteristic value of the zero sequence current of each feeder terminal under each data window;
the characteristic values comprise three types of characteristic values of skewness, variance and extreme values, and each feeder terminal is provided with D sampling points in P + Q power frequency periods;
and step 3: constructing a characteristic value curve of each feeder terminal according to the characteristic values calculated in the step 2;
the characteristic value curve comprises a skewness curve, a variance curve and an extreme value curve, and the characteristic value curve is a characteristic value-sampling point curve;
sampling points in the characteristic value curve are sampling points in the first P + Q-1 power frequency cycles in the zero sequence current of the feeder terminal extracted in the step 1;
and 4, step 4: acquiring a sampling point corresponding to the maximum characteristic value in each characteristic value curve of the feeder terminal closest to the bus on the fault feeder;
and 5: obtaining various characteristic values corresponding to the same sampling points in the step 4 in characteristic value curves of the rest M-1 feeder terminals on the fault feeder;
and 6: sequentially calculating Euclidean distances of the characteristic values of two adjacent feeder line terminals according to the maximum characteristic value in the step 4 and the characteristic value obtained in the step 5;
and 7: arranging the Euclidean distances calculated in the step 6 from large to small, and selecting the maximum N Euclidean distances from the Euclidean distances;
wherein N is more than or equal to 3 and less than or equal to M-1;
and step 8: if the Euclidean distance of the characteristic value of one adjacent two feeder terminals is larger than the sum of the remaining N-1 Euclidean distances in the step 7, the interval between the two adjacent feeder terminals corresponding to the Euclidean distance of the characteristic value of the one adjacent two feeder terminals is a fault interval; and if the fault locating device does not exist, returning to the step 1 to reselect the fault feeder line for fault location.
2. The method according to claim 1, wherein the deviation calculation formula in the feature value in step 2 is as follows:
wherein, sk (i) 0ab (j) Represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab Skewness of (d);
representing a random variable ofThe desired operator of (a);
i 0ab (n) is the zero sequence current sampled by the feeder terminal with number b on the fault feeder a at the sampling point n, n is the sampling point, ns is the number of the sampling points under the selected data window length, mu and sigma respectively represent i 0ab Expectation and standard deviation of.
3. The method according to claim 1, wherein the variance calculation formula in the feature value in step 2 is as follows:
Va(i 0ab (j))＝E[(i 0ab (n)-E(i 0ab (n))) 2 ],n＝j,j+1,...,j+Ns-1
wherein, va (i) 0ab (j) Represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab The variance of (a);
E[(i 0ab (n)-E(i 0ab (n))) 2 ]represents a random variable of (i) 0ab (n)-E(i 0ab (n))) 2 The desired operator of (a); e (i) 0ab (n)) represents a random variable of i 0ab (n) a desired operator;
i 0ab and (n) is zero sequence current collected by a feeder terminal with the number b on the fault feeder a at a sampling point n, wherein n is the sampling point, and Ns is the number of sampling points under the selected data window length.
4. The method according to claim 1, wherein the extreme value calculation formula in the feature value in step 2 is as follows:
Ra(i 0ab (j))＝max(i 0ab (n))-min(i 0ab (n)),n＝j,j+1,...,j+Ns-1
wherein, ra (i) 0ab (j) Represents the zero-sequence current i in the data window with the jth sampling point as the starting point 0ab An extreme value of (d);
max(i 0ab (n)) represents the zero sequence current i in the data window containing Ns sampling points 0ab Maximum value of (d);
min(i 0ab (n)) represents the zero sequence current i in the data window containing Ns sampling points 0ab N is the number of samples, and Ns is the number of samples at the selected data window length.
5. The method according to claim 1, wherein the euclidean distance between two adjacent feeder terminal characteristic values in step 6 is calculated as follows:
wherein d (k, k-1) represents the Euclidean distance of characteristic values of the kth feeder terminal and the kth-1 feeder terminal;
Sk k 、Va k 、Ra k respectively representing a skewness curve and a variance curve of the kth feeder terminal and a skewness, a variance and an extreme value corresponding to the same sampling point in the step 4 on an extreme value curve; sk k-1 、Va k-1 、Ra k-1 And 4, respectively representing the skewness, the variance and the extreme value corresponding to the same sampling point in the step 4 on the skewness curve, the variance curve and the extreme value curve of the k-1 th feeder terminal.
6. The method of claim 1, wherein N of the maximum N euclidean distances selected in step 7 is 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710687706.3A CN107632225B (en) | 2017-08-11 | 2017-08-11 | A kind of small current system Earth design method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710687706.3A CN107632225B (en) | 2017-08-11 | 2017-08-11 | A kind of small current system Earth design method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107632225A true CN107632225A (en) | 2018-01-26 |
| CN107632225B CN107632225B (en) | 2019-07-23 |
Family
ID=61099241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710687706.3A Active CN107632225B (en) | 2017-08-11 | 2017-08-11 | A kind of small current system Earth design method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107632225B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108981755A (en) * | 2018-08-16 | 2018-12-11 | 国电南瑞科技股份有限公司 | Rotor earth fault protection DC quantity mean value calculation data selecting method |
| CN109507529A (en) * | 2018-11-16 | 2019-03-22 | 威胜信息技术股份有限公司 | Small current grounding fault based on fault detector differentiates localization method |
| CN111736038A (en) * | 2020-08-18 | 2020-10-02 | 广东冠能电力科技发展有限公司 | Edge calculating device for power distribution network fault positioning and power distribution network fault positioning system |
| CN111830365A (en) * | 2020-06-20 | 2020-10-27 | 青岛鼎信通讯股份有限公司 | Multi-transient statistical fault positioning method applied to power grid fault diagnosis |
| CN112067948A (en) * | 2020-10-14 | 2020-12-11 | 长沙理工大学 | Fault line selection method, system and terminal for single-phase earth fault of power distribution network and readable storage medium |
| CN112067884A (en) * | 2020-09-16 | 2020-12-11 | 广东电网有限责任公司 | Feeder branch fault detection method and system |
| CN113625125A (en) * | 2021-09-24 | 2021-11-09 | 南方电网科学研究院有限责任公司 | High-resistance ground fault detection method, device and equipment for power distribution network |
| CN115356658A (en) * | 2022-10-21 | 2022-11-18 | 国网信息通信产业集团有限公司 | Power distribution network single-phase earth fault starting detection method and system |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004233255A (en) * | 2003-01-31 | 2004-08-19 | Mitsubishi Electric Corp | Disconnection detecting system of distribution line |
| CN101598761A (en) * | 2009-07-29 | 2009-12-09 | 江苏省电力公司常州供电公司 | Fault line selection method for small electric current grounding system of distribution network |
| CN102253293A (en) * | 2011-05-30 | 2011-11-23 | 南京航空航天大学 | Detection method and device of direct current arc faults |
| CN103091603A (en) * | 2013-01-14 | 2013-05-08 | 华北电力大学 | Breakdown intelligent classification and positioning method of electric transmission line |
| CN103267926A (en) * | 2013-03-19 | 2013-08-28 | 中国石油大学(华东) | Data-gram (DG)-containing power distribution network fault distance measurement for fault feature matching based on differential evolution algorithm |
| CN103308822A (en) * | 2013-05-07 | 2013-09-18 | 河南理工大学 | Small current earth fault line selection method for radial distribution network |
| CN103529356A (en) * | 2012-07-03 | 2014-01-22 | Abb技术有限公司 | Method and apparatus for determining the distance to phase-to-earth fault |
| CN103675605A (en) * | 2013-12-11 | 2014-03-26 | 湖南大学 | Small-current earth fault line selection method based on fault signal transient state correlation analysis |
| CN103837784A (en) * | 2014-03-27 | 2014-06-04 | 国家电网公司 | Artificial intelligence identification method of external faults of conversion current bus of HVDC system |
| CN105866669A (en) * | 2016-05-19 | 2016-08-17 | 华南理工大学 | Circuit breaker breaking-closing control loop fault diagnosis method |
| WO2017002020A1 (en) * | 2015-06-29 | 2017-01-05 | Abb Schweiz Ag | System and method for detecting a fault in an overhead power line |
| CN106526410A (en) * | 2016-09-07 | 2017-03-22 | 中国石油大学(华东) | Small-current grounding system high-resistance grounding fault positioning method based on transient current projection component amplitude comparison |
| CN106990330A (en) * | 2017-05-27 | 2017-07-28 | 国家电网公司 | The discrimination method of one-phase earthing failure in electric distribution network phase |
| CN106990324A (en) * | 2017-04-06 | 2017-07-28 | 北京四方继保自动化股份有限公司 | A kind of distribution net work earthing fault detects localization method |
-
2017
- 2017-08-11 CN CN201710687706.3A patent/CN107632225B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004233255A (en) * | 2003-01-31 | 2004-08-19 | Mitsubishi Electric Corp | Disconnection detecting system of distribution line |
| CN101598761A (en) * | 2009-07-29 | 2009-12-09 | 江苏省电力公司常州供电公司 | Fault line selection method for small electric current grounding system of distribution network |
| CN102253293A (en) * | 2011-05-30 | 2011-11-23 | 南京航空航天大学 | Detection method and device of direct current arc faults |
| CN103529356A (en) * | 2012-07-03 | 2014-01-22 | Abb技术有限公司 | Method and apparatus for determining the distance to phase-to-earth fault |
| CN103091603A (en) * | 2013-01-14 | 2013-05-08 | 华北电力大学 | Breakdown intelligent classification and positioning method of electric transmission line |
| CN103267926A (en) * | 2013-03-19 | 2013-08-28 | 中国石油大学(华东) | Data-gram (DG)-containing power distribution network fault distance measurement for fault feature matching based on differential evolution algorithm |
| CN103308822A (en) * | 2013-05-07 | 2013-09-18 | 河南理工大学 | Small current earth fault line selection method for radial distribution network |
| CN103675605A (en) * | 2013-12-11 | 2014-03-26 | 湖南大学 | Small-current earth fault line selection method based on fault signal transient state correlation analysis |
| CN103837784A (en) * | 2014-03-27 | 2014-06-04 | 国家电网公司 | Artificial intelligence identification method of external faults of conversion current bus of HVDC system |
| WO2017002020A1 (en) * | 2015-06-29 | 2017-01-05 | Abb Schweiz Ag | System and method for detecting a fault in an overhead power line |
| CN105866669A (en) * | 2016-05-19 | 2016-08-17 | 华南理工大学 | Circuit breaker breaking-closing control loop fault diagnosis method |
| CN106526410A (en) * | 2016-09-07 | 2017-03-22 | 中国石油大学(华东) | Small-current grounding system high-resistance grounding fault positioning method based on transient current projection component amplitude comparison |
| CN106990324A (en) * | 2017-04-06 | 2017-07-28 | 北京四方继保自动化股份有限公司 | A kind of distribution net work earthing fault detects localization method |
| CN106990330A (en) * | 2017-05-27 | 2017-07-28 | 国家电网公司 | The discrimination method of one-phase earthing failure in electric distribution network phase |
Non-Patent Citations (2)
| Title |
|---|
| 刘鹏辉 等: "基于动态时间弯曲距离的小电流接地故障区段定位方法", 《电网技术》 * |
| 戴栩生 等: "基于暂态相关性分析的小电流接地故障选线方法", 《中国电力》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108981755B (en) * | 2018-08-16 | 2020-09-29 | 国电南瑞科技股份有限公司 | Method for selecting calculation data of average value of Direct Current (DC) quantity of rotor grounding protection |
| CN108981755A (en) * | 2018-08-16 | 2018-12-11 | 国电南瑞科技股份有限公司 | Rotor earth fault protection DC quantity mean value calculation data selecting method |
| CN109507529A (en) * | 2018-11-16 | 2019-03-22 | 威胜信息技术股份有限公司 | Small current grounding fault based on fault detector differentiates localization method |
| CN109507529B (en) * | 2018-11-16 | 2020-11-13 | 威胜信息技术股份有限公司 | Small current grounding fault distinguishing and positioning method based on fault indicator |
| CN111830365A (en) * | 2020-06-20 | 2020-10-27 | 青岛鼎信通讯股份有限公司 | Multi-transient statistical fault positioning method applied to power grid fault diagnosis |
| CN111736038A (en) * | 2020-08-18 | 2020-10-02 | 广东冠能电力科技发展有限公司 | Edge calculating device for power distribution network fault positioning and power distribution network fault positioning system |
| CN112067884A (en) * | 2020-09-16 | 2020-12-11 | 广东电网有限责任公司 | Feeder branch fault detection method and system |
| CN112067948A (en) * | 2020-10-14 | 2020-12-11 | 长沙理工大学 | Fault line selection method, system and terminal for single-phase earth fault of power distribution network and readable storage medium |
| CN112067948B (en) * | 2020-10-14 | 2022-03-11 | 长沙理工大学 | Fault line selection method and system for single-phase earth fault of power distribution network |
| CN113625125A (en) * | 2021-09-24 | 2021-11-09 | 南方电网科学研究院有限责任公司 | High-resistance ground fault detection method, device and equipment for power distribution network |
| CN113625125B (en) * | 2021-09-24 | 2023-11-21 | 南方电网科学研究院有限责任公司 | High-resistance ground fault detection method, device and equipment for power distribution network |
| CN115356658A (en) * | 2022-10-21 | 2022-11-18 | 国网信息通信产业集团有限公司 | Power distribution network single-phase earth fault starting detection method and system |
| CN115356658B (en) * | 2022-10-21 | 2023-03-07 | 国网信息通信产业集团有限公司 | Power distribution network single-phase earth fault starting detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107632225B (en) | 2019-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107632225A (en) | A kind of small current system Earth design method | |
| CN108181547B (en) | Dynamic time bending distance fault section positioning method based on time sequence compression | |
| CN110082640B (en) | Distribution network single-phase earth fault identification method based on long-time memory network | |
| CN103344875B (en) | Classification line selection method for single-phase earth fault of resonance earthing system | |
| CN109307824B (en) | Clustering-based power distribution network single-phase earth fault section positioning method | |
| CN111313932B (en) | Transmission line based on electromagnetic time reversal transfer function correlation and network fault positioning method | |
| CN108448540B (en) | Zero sequence current comparison-based ground fault protection method for small-resistance grounding system | |
| CN108508320B (en) | Arc grounding fault identification method based on harmonic energy and wave distortion feature | |
| CN103941156B (en) | Multi-information acquisition Section Location based on extreme learning machine | |
| CN106990324B (en) | Power distribution network ground fault detection and positioning method | |
| CN107045093B (en) | Low-current single-phase earth fault line selection method based on quick S-transformation | |
| CN112684279B (en) | Phase current similarity-based power distribution network single-phase earth fault detection algorithm | |
| CN106093705B (en) | A kind of calculation method of one-phase earthing failure in electric distribution network wavefront | |
| CN110247420B (en) | Intelligent fault identification method for HVDC transmission line | |
| CN112557812B (en) | Small current ground fault positioning method and system based on Hausdorff distance | |
| CN102135588B (en) | Intelligent ranging method for radiation network fault by utilizing S transformation spectrum characteristics | |
| CN117783729A (en) | Lightning stroke interference identification method and system for traveling wave transient protection | |
| CN107632237A (en) | A kind of fault line selection method for single-phase-to-ground fault based on wavelet character amount correlation after failure | |
| Aslan et al. | ANN based fault location for medium voltage distribution lines with remote-end source | |
| CN107632235B (en) | Device and method for identifying fault line of low-resistance grounding system | |
| CN115754565A (en) | Method and system for detecting potential safety hazard of electric bicycle during charging | |
| Sebastian et al. | Location of fault in a dc microgrid using state space model based approach | |
| Zhu et al. | High impedance grounding fault detection in resonance grounding system based on nonlinear distortion of zero-sequence current | |
| CN113358972A (en) | High-resistance ground fault line selection method based on line transient characteristics | |
| Yang et al. | Multi-criteria Fusion Method for Line Selection of Single-phase-to-ground Fault with Small Current based on GRU Neural Network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |